DE60308020T2 - THREE-DIMENSIONAL INSULATED POSITION SENSOR - Google Patents

Description

BACKGROUND THE INVENTION

The The present invention relates to vehicle occupant safety systems and more particularly, a vehicle occupant proximity sensor for use with a vehicle occupant safety system.

Vehicle occupant safety systems which are activated in response to a vehicle accident Injuries to occupants are in the technical field well known. A vehicle can have automatic safety retention actuators include, for example, front and side airbags, seatbelt pretensioners and deployable knee pads. The occupant protection system may also a collision / accident sensor to detect the occurrence of a Vehicle accident or vehicle collision and an electrical Provide signal indicative of the magnitude of the accident.

Several known occupant protection systems include an occupant position sensor, the position of the occupant with respect to an associated inflatable or unfoldable protection module detected. The occupant position sensor for a such a system could an ultrasonic sensor, an infrared sensor, and a capacitive sensor, and / or weight sensor. A controller with the sensors connected, controls the inflatable protection module in response on the detected position of the occupant. In response to the detected occupant position One or more deployment aspects of the airbag may be adjusted become. A protection system with adjustable aspects for deployment usually becomes referred to as an "adaptive" protection system. When the occupant (vehicle passenger) positions at a position is, so that the deployment of the airbag provides protection to the occupant will not improve, it may be desirable its a trigger of the occupant protection module. An inmate who is very close to the protection module is referred to as being within an out-of-position Occupant zone is. The deployment of the airbag for an occupant within the out-of-position Occupant zone can not improve protection of the occupant.

In In any case, determining the position of the occupant is more important Part of the adaptive occupant protection system. There are several types of proximity sensors, such as an ultrasonic sensor, a video sensor, a capacitive Sensor and an infrared sensor. Different obstacles, like for example, a map, a book or a newspaper, could be signals from ultrasound and Mask video sensors. A cigarette lighter or a cigarette could be one Dim infrared sensor.

The US 2001/0003168 relates to an arrangement in a vehicle for determining the vehicle occupant position relative to a fixed one Construction within the vehicle with an array, which is an image a portion of the vehicle passenger compartment for which an arrangement of the occupant probably, and with a lens that is between the array and the part of the vehicle passenger compartment. The picture can be adjusted by adjusting the lens, z. B. by adjusting the focal length the lens and / or the position of the lens relative to the array, by adjusting the array, e.g. B. the position of the array relative to the lens, and / or by using software to perform a Focusing process changed become. By changing of the image to determine the clearest image, d. H. to focus of the image, can be a distance between the occupant and the solid Structure can be obtained on the basis of the parameters of the change of the image.

The DE 19841399 relates to a user status determination arrangement for a vehicle seat that is used to determine whether a seat is occupied to control an airbag. The assembly has means for measuring the distance from an upper defined point to a plurality of defined points on and around the seat. Contour information is derived from a photograph of the area having at least a portion of the seat. The assignment is determined from the distance information and the derived contour.

The Invention is based on conductivity of the human body supported. This phenomenon allows the occupant to be used as a transmitting antenna, being his / her position within a defined space through a measurement of electromagnetic values that are at a rate of recipients be determined is determined.

SUMMARY THE INVENTION

• (a) Messen eines elektromagnetischen Felds, dass durch einen Insassen hervorgerufen wird, an einer Vielzahl von Punkten in einem Fahrzeugpassagierabteil; und
• (b) Berechnen einer teildimensionalen Position des Insassen in dem Passagierabteil unter Verwendung einer Signalverteilung auf Grundlage des Schritts (a).

According to a first aspect of the present invention, a method for determining a Po tion of a vehicle occupant, comprising the following steps:

• (a) measuring an electromagnetic field caused by an occupant at a plurality of points in a vehicle passenger compartment; and
• (b) calculating a partial position of the occupant in the passenger compartment using a signal distribution based on the step (a).

The present invention provides an occupant position sensor the conductivity An occupant uses the height and position of the occupant to determine by the capacity between the head of the occupant and a variety of mounted on the roof Sensors (electrodes) is measured.

A Transmitting electrode is mounted in a vehicle seat. An arrangement of receiving electrodes is on the roof of the vehicle above the Seat of the occupant attached. The sensor uses the conductivity of the human body by using the occupant as a sender. The highest point of the body an inmate is considered a source for considered the electromagnetic waveform. The values of the signals, the be evoked on each electrode are controlled by the control unit measured and then processed to determine the position and the movement of the highest Point of the body to track an inmate. The procedure that in this Invention provides the possibility of a three-dimensional Position of the highest Point of the body of the occupant within the detection area.

SUMMARY THE DRAWINGS

Other Advantages of the present invention are readily apparent as well as this better with reference to the following detailed Description understandable when considered in conjunction with the accompanying drawings becomes. In the drawings show:

1 the vehicle occupant proximity sensor installed in a vehicle passenger compartment having an occupant protection system;

2 a schematic representation illustrating the calculation of the position and height of the occupant;

3 another schematic diagram illustrating the method for calculating the position and height of the occupant;

4 a graph illustrating the linear interpolation between adjacent samples of the autocorrelation function;

5a a graph of a source signal;

5b Graphs of functions f1 (K ₁₎ (line 1), f2 _(Ki) (line 2), f1 (K _i) f2 _(Ki) (line 3) which were calculated for the source signal as shown in 5a shown;

6a a graph of a source signal in an example when two objects generate a resolution element in a signal;

6b a graph of the autocorrelation function for the source signal of 6b ; and

6c a graph of f1 (line 1), f2 (line 2) and f1-f2 (line 4) for the source signal for the 6a ,

DETAILED DESCRIPTION OF A PREFERRED EMBODIMENT

1 illustrates a vehicle occupant proximity sensor 10 for determining the height and position of an occupant 12 in a vehicle seat 14 and in particular for determining the three-dimensional position of the head 15 of the occupant. The inmate 12 and the vehicle seat 14 are in a vehicle passenger compartment 16 with an occupant protection system including an automatic safety restraint such as an air bag 18 , Installed. Although an airbag mounted on the steering wheel 18 As an example, it should also be appreciated that the present invention is also useful for side airbags, seatbelt pretensioners, deployable knee bolsters, and any other automatic safety restraint actuators. The accident detector 19 , such as an accident sensor of any known type, ver used to determine the occurrence of a vehicle accident and determine the accident severity.

The vehicle occupant proximity sensor 10 includes a transmission electronics 20 which generates an electromagnetic signal, and a first array (field) 22 of receiving electrodes 22a -n, perpendicular to a second field 23 of receiving electrodes 23a -n cut. The receiving electrodes 22a -n, 23a -n receive the electromagnetic signal transmitted through the transmitting electrode 20 is produced. A control unit 24 receives electrical signals from the receiving electrodes 22a -n, 23a -n based on the electromagnetic signal that passes through the electrodes 22a -n, 23a -n is received. The control unit 24 may also be a signal from a seat rail position sensor 26 receive the position of the vehicle seat 14 on a vehicle rail (not shown) in the passenger compartment 16 displays.

The transmitting electrode 20 is in the base of the vehicle seat 14 appropriate. The transmitting electrode 20 may comprise a coil of wire or a copper layer and may be formed of any conductive material, but preferably includes a grid of copper wires spaced about one inch apart. Generally, a large area of the base of the seat is preferred 14 with the transmitting electrode 20 cover and wrap the transmitting electrode around the front part of the seat. It should be ensured that the transmitting electrode is not shorted to ground across the frame of the vehicle. A frequency generator 27 generates a low frequency, low amplitude signal, such as a 10 KHz signal, at the transmitting electrode 20 , which then serves as an electromagnetic signal in the passenger compartment 16 is sent.

The receiving electrode arrays 22 . 23 are in the vehicle roof paneling 28 in the passenger compartment 16 over the occupant 12 appropriate. The receiving electrodes 22a -n, 23a -n each include a small conductive surface, preferably a 6.5 cm x 9 cm piece of printed circuit board, but the size will vary depending on the passenger compartment size. The conductivity can be achieved by a suitable choice of the substance, the foil, the grid, a conductive paint or any other conductive material. The receiving electrodes 22a -n, 23a -n are with the control unit 24 via a multiplexer 29 and an amplifier 30 connected. Again, it must be ensured that none of the receiving electrodes 22a -n, 23a -n over the frame are shorted to ground. The multiplexer 29 allows the control unit 24 sequentially values from the receiving electrodes 22a -n, 23a -n to read the three-dimensional position of the occupant 12 in accordance with the procedure described below. Analog-to-digital converters (not shown) would receive the signals from the amplifiers 30 convert to a computer-readable format.

The control unit 24 generally includes a CPU 31 with a memory 32 , for example, a RAM, ROM, DVD, CD, hard disk drive, or other electronic, optical, magnetic, or other computer readable medium having stored therein programs for carrying out the steps and algorithms described herein. The CPU 31 is suitably programmed to perform the functions described herein, and one of ordinary skill in the art could use the CPU 31 and provide any additional hardware not shown but needed to implement the present invention based on the description herein. The power supply and additional hardware are not shown, but those of ordinary skill in the art could implement the present invention based on the description herein.

In operation, the control unit controls 24 the generator 27 to get a 10 KHz signal at the transmitting electrode 20 to create. The transmitting electrode 20 sends a 10 KHz signal as an electromagnetic wave inside the vehicle passenger compartment 16 , The electromagnetic signal passes through the occupant 12 and is from the receiving electrodes 22a -n, 23a -n received. The signal received by each receiving electrode is the capacitance between it and the transmitting electrode 20 in turn, depending on the proximity of the occupant 12 to each receiving electrode 22a -n, 23a -n will change. The size, spacing and number of electrodes in each of the receiving electrode arrays 22 . 23 may differ for different applications and different vehicles.

The control unit 24 controls a multiplexer 29 to sequentially each of the receiving electrodes 22a -n, 23a -n in the arrays 22 . 23 read. Although they are read out sequentially, this becomes sufficiently fast relative to a normal movement of a vehicle occupant 12 to effectively provide a current snapshot of sufficient information about the height and position of the occupant 12 in the passenger compartment 16 to determine. The capacity at each receiving electrode 22a -n, 23a -n depends on the proximity of the occupant 12 to each receiving electrode 22a -n, 23a -n. Thus, the highest capacity measured at the receiving electrode closest to the head 15 of the vehicle occupant 12 is.

The occupant is coupled to the oscillator via the seat-mounted electrode to provide a more capacitive connection. The electromagnetic wave is transmitted from the highest point of the occupant and caused to each electrode. The value of the signal on each electrode is a function of the distance between the first point of the occupant and the electrode (see 2 ). amplifier 30 send these values to the control unit 24 ,

Referring to the simplified illustration, which in FIG 2 is shown, the values of the signal are processed to each electrode independently for each line to determine the x (or y) position coordinates along the line and the vertical distance h _x (or h _y ) to each line. The values are then used to calculate 3D coordinates in accordance with the following equation: H 2 = I 2 x - y 2 = h 2 y - x 2 (1)

2D coordinates for each line are determined by using a waveform according to the following considerations:
The electric field in the row is determined by measuring the electric field produced on the flat electrode. The linear size of the electrode and the typical size of the object are much larger than adequate accuracy. Therefore, the determination of an absolute occupant position is solved using the calibration approach. During the calibration routine, both the signal from a typical object used as a calibration signal and the absolute position of the object in memory 32 saved. During operation, the position of a real object is determined as a displacement of the object signal relative to the calibration signal.

The electric field of a complex object is an overlay of point sources of charge, so that the current system is a linear displacement-invariant System for the given height is. A convolution of a current signal with a stored calibration signal is used to determine the position along the line.

wobei S_m eine Sequenz des Signalabtastwerts, C_n – eine Kalibrierungssignalsequenz, und Ψ und Ψ^–1 ein Paar von diskreten Fourier-Transformationen ist, definiert folgendermaßen:

The position of the convolution maximum corresponds to the value of the displacement of the present signal from the calibration signal. To avoid calibration requirements for each altitude, a convolution is used with the calibration signal with the mean altitude value. A convolution K _n is calculated by a Fourier transformation in accordance with the convolution theorem:

where S _{m is} a sequence of the signal sample, C _n - a calibration signal sequence, and Ψ and Ψ ^{-1 is} a pair of discrete Fourier transforms, defined as follows:

One Algorithm for a fast Fourier transform (FFT) is used to get the calculate discrete Fourier transform.

For a more accurate determination of the convolution maximum position, a Fourier interpolation is used during the calculation of the inverse FFT, wherein the frequency domain is widened by being padded with zeros. In accordance with the theory of linear systems, this gives a more accurate construction of the continuous sequence if the source sequence was sampled with an interval Δ = 1 / 2f _c , where f _{c is} the Nyquist critical frequency. Further, the condition of non-overlapping frequency spectrums of the signal and the sampling function was maintained. In this case, the continuous sequence S (x) is given by the following formula:

Certain weighting functions, such as Hamming or Kaiser functions, are typically used to reduce the effect of spectrum overlap [2]. In the present situation, it seems impossible to use the weighting method for 8 samples of a source sequence. Instead, the source sequence is padded with zeros up to 16 samples, so that it periodically becomes period 16. After the forward FFT, the frequency domain is padded with zeros up to 64. Multiplication of the sequence by Ψ (C _n ) and calculation of the inverse FFT results in 32 interpolation samples of the convolution K _n in accordance with equations (2-4).

wobei q ein elektrischer Ladungswert ist, r der Abstand zwischen der Ladung und einem Messpunkt ist, und a eine Konstante ist.The system and method of the present invention also determine the altitude coordinate. A signal generated by a point electric charge is referred to as a "simple wave." The point charge generates an electric field:

where q is an electric charge value, r is the distance between the charge and a measurement point, and a is a constant.

Referring to 3 For example, an electric field is measured along a line placed at a distance h from the charge so that the point near the charge point is located at x = 0. Then the signal distribution along the line will be as follows:

The signal at the point x = 0 is:

A calculation of the quotient S (x) / S (0) leads to the following equation:

Out of equation (9), the value of h can be evaluated. The precision the evaluation depends strong from the definition of the x coordinate, the position of the maximum and the degree of signal noise. An increase in the number of receptors can lead to a reduction in the receptor area and thus reduce the signal-to-noise ratio. Thus, a signal between sample points should be used be restored to an interpolation.

It There are different procedures for a signal recovery. However, these differ by a computational complexity. In accordance with the form of the present source function may be the best method be a polynomial approximation. However, with good results Accuracy can generate this method a complex calculation, like a decomposition of a singular Require matrix. It is proposed to use a Fourier analysis a Fourier interpolation of the signal between sample points for a more accurate one Definition of the source coordinate to use.

Consider the autocorrelation K (x) = S (x) ⨂S (x) (the convolution S (x) * S (x) is the same because of the symmetric S (x)):

was wie (8) mit H = 2h aussiehtThe expression for the quotient K (x) / K (0) is:

what looks like (8) with H = 2h

If you compare (10) with (11) you get:

So can be concluded that h not only using the signal, but also evaluated by the signal autocorrelation function can be. This function hangs from an integral characteristic of the signal and thus is across from Noise is less sensitive. Additionally, this feature is theirs Kind after symmetric and has a maximum exactly at x equal to zero. Thus, the problem of an exact maximum position definition is not a present one questionable aspect. Experiments have shown that the equation (12) for I work well. However, a weak dependence was found, namely the h from the level of hardware gain depends for themselves different K (x) parts is different. For a better understanding will be performed the following analysis of the K (x) form.

Define inverse functions H (K) and x (K) so that the argument k increases monotonically from 0 to Kmax = K (0), ie 0 ≤ K ≤ Kmax (13) if H (K) and x (K) (12) satisfy, one obtains:

H (K) should be a constant and therefore its derivative is zero. Thus one obtains: H (K)

With a replacement of δH / δK = 0 one obtains from (16):

The following functions are calculated for the signal under investigation:

The points at which f1 (K) and f2 (K) intersect will be the best match for determining h by using Equation (14-15) or (9). In an ideal situation, f1 (K) must be exactly the same as f2 (K), but because of the sampling, computation, and deviation from the real waveform and the ideal l / r ² function, f1 (K) and f2 (K) do not become equal be. This can be improved by the presence of more than one crossover point by using an average of h values calculated at those points.

The autocorrelation function K _{n of} the source signal sequence S _n is calculated by using the FFT algorithm according to Equations (2-3). S _n is padded with zeros up to 16 samples and Fourier interpolation is used to obtain 32 samples of the autocorrelation function.

wobei N = 32 ist.Discrete functions f1 (K _i ) and f2 (K _i ) are set up (see 4 ) using a linear interpolation between adjacent samples of K _n :

where N = 32.

die aus (15) und (17) ermittelt wird, wenn H = 2 h ist.:, F2 (K _i) → 0 holds calculated using the following equation - in the next step h is at the points where f1 (K _i)

which is determined from (15) and (17) when H = 2 h.

5b shows functions f1 (K _i ) (line 1), f2 (K _i ) (line 2), f1 (K _i ) -f2 (K _i ) (line 3) calculated for a source signal, as in FIG 5a shown.

6 shows an example when two objects generate a resolution element in a signal. 6a is a source signal, 6b is their autocorrelation function, 6c shows f1 (line 1), f2 (line 2) and f1-f2 (line 4).

This Algorithm includes a 16-point FFT, a 64-point FFT, a 32-point FFT and a few vector operations, such as multiplication, Subtraction and a maximum search to the 2D position for each line to determine.

The control unit 24 monitors the information from the receiving electrode array 22 over time. For example, the position of the head may be 15 of the occupant 12 do not change immediately; she has to follow a path from one point to another. The control unit 24 may additionally provide information from the vehicle seat rail sensor 26 use the position of the vehicle seat 14 displayed on a vehicle seat rail.

All information is provided by the control unit 24 used to determine whether based on an accident by the accident detector 19 is detected, and the strength of an accident to determine if the airbag 18 (or another security restraint device) is to be deployed. If, for example, the control unit 24 based on information from the receiving electrode array 22 it is determined that the inmate 12 too close to the airbag 18 is, then the control unit 24 determine, in the event of an accident, the airbag 18 not to activate, or the control unit 24 can determine that the airbag 18 should be deployed with less force. On the other hand, if the control unit 24 of the occupant 12 based on information from receiving electrode arrays 22 . 23 determines that the occupant is at a distance from the airbag 18 is above a predetermined threshold, becomes the control unit 24 the airbag 18 to unfold or is a deployment of the airbag 18 effect with higher force, depending on the severity of the accident, as by the accident detector 19 certainly. The control unit 24 Also determines the force with which the airbag 18 (or another safety restraint) should be deployed based on the height of the occupant 12 ,

Additionally, information may be from the seat rail sensor 26 may be used with the approximate information to determine if and / or how the airbag 18 should be inflated or unfolded. If, for example, the seat rail sensor 26 indicates that the vehicle seat 14 in the vehicle passenger compartment 16 is set to the front and the receiving electrode arrays 22 . 23 show that the inmate 12 also front, the control unit can 24 determine the airbag 18 in the event of an accident unfold. On the other hand, when the seat rail position sensor indicates that the vehicle seat 14 too far forward, the control unit can 24 decide the airbag 18 not unfold, although the receive electrode arrays 22 . 23 show that the head 15 of the occupant 12 is sufficiently far to the rear for deployment. This would happen in the event that the inmate 12 the vehicle seat 14 has set very obliquely. Furthermore, the control unit 24 determine that when the head 15 of the occupant 12 strongly tilted backwards, the airbag 18 is deployed in the event of an accident, although the vehicle seat rail position sensor 26 indicates that the vehicle seat 14 is too far forward. This would indicate that the inmate 12 turn the vehicle seat 14 strong and sufficiently inclined that the airbag 18 should be deployed. In general, those of ordinary skill in the art will become the control unit 24 using the above and numerous additional rules about whether the airbag 18 is to trigger, and for a multi-stage airbag 18 with how much power the airbag 18 should be deployed, program. The present invention additionally provides information to the control unit 24 so that those of ordinary skill in the art could use this additional information to determine in a suitable manner whether and with how much force the airbag 18 is to be activated.

In accordance with the requirements of patent laws and case-law, exemplary configurations described above are considered to illustrate a preferred embodiment of the invention. However, it should be understood that the invention has been implemented in a different manner can be considered specific and described without departing from its scope. The invention is defined by the appended claims.

Claims (15)

Method for determining a position of a vehicle occupant ( 12 ), comprising the steps of: (a) measuring an electromagnetic field detected by an occupant ( 12 ) at a plurality of points in a vehicle passenger compartment ( 16 ); and (b) calculating a three-dimensional position of the occupant ( 12 ) in the passenger compartment ( 16 ) using a signal distribution based on the step (a). The method of claim 1, wherein step (b) the step of using a signal distribution function, to calculate the position includes. The method of claim 1, wherein step (b) the step of comparing the electromagnetic field, which is caused at a first of the plurality of points, with the electromagnetic field attached to a second of the plurality caused by points we include. The method of claim 3, wherein step (a) further comprises the step of transmitting an electromagnetic signal from a vehicle seat (10). 14 ) in the vehicle passenger compartment ( 16 ). The method of claim 4, wherein the plurality of Points are arranged substantially in one plane. The method of claim 1, further comprising the steps of: storing a calibration signal; and determining the position of the occupant ( 12 ) in the step (b) based on a comparison of the electromagnetic field measured in the step (a) with the calibration signal. Vehicle passenger compartment proximity detection system ( 10 ), comprising: a plurality of sensors having a capacity at a plurality of points in a passenger compartment ( 16 ), the measurements at the plurality of points forming a signal distribution; and a controller ( 24 ), which is a three-dimensional position of an occupant ( 12 ) using the signal distribution based on the capacitance measured at the plurality of points. Vehicle passenger compartment proximity detection system ( 10 ) according to claim 7, further comprising: a first electrode ( 20 ) which generates an electromagnetic signal; wherein the capacitance sensors comprise a plurality of second electrodes ( 22a -n, 23a -n) receiving the electromagnetic signal; and where the controller ( 24 ) determines the capacitance at each of the second electrodes to determine the three-dimensional position of the occupant ( 12 ) in the vehicle passenger compartment ( 16 ). Vehicle passenger compartment proximity detection system ( 10 ) according to claim 8, wherein the first electrode ( 20 ) in a vehicle seat ( 14 ) in the vehicle passenger compartment ( 16 ) is attached. Vehicle passenger compartment proximity detection system ( 10 ) according to claim 9, wherein the plurality of second electrodes ( 22a -n, 23a -n) adjacent to a vehicle roof inner lining in the vehicle passenger compartment ( 16 ) is attached. A computer readable medium containing a computer program that, when executed by a computer, performs the steps of: (a) receiving a plurality of measurements of an electromagnetic field detected by an occupant ( 12 ) at a plurality of points in a vehicle passenger compartment ( 16 ); and (b) calculating a three-dimensional position of the occupant ( 12 ) in the passenger compartment ( 16 ) using a signal distribution based on the plurality of measurements. A computer readable medium as claimed 11, wherein the step (b) comprises the step of using a signal distribution function, to calculate the position includes. The computer-readable medium of claim 12, wherein step (b) comprises the step of ver equal to differences under the electromagnetic field induced at the plurality of points. A computer readable medium as claimed 13, wherein the plurality of points are assumed to be are arranged substantially in one plane. The computer-readable medium of claim 11, further comprising the step of: determining the position of the occupant ( 12 ) in the step (b) based on a comparison of the plurality of measurements with a calibration signal.